Deep Dives
Deep Dives
Google’s Vantage Experiment: Using GenAI to Score ‘Future-Ready’ Skills
Google Research and NYU built Vantage, an AI-powered assessment tool that scores critical thinking, collaboration,...
Deep Dives
ConvApparel: Closing the Realism Gap in AI User Simulators
Google Research's ConvApparel dataset and evaluation framework measure how well LLM-based user simulators mimic real...
Deep Dives
Simula: Google’s New Framework for Building Synthetic Datasets That Actually Work
Google Research introduces Simula, a reasoning-first framework that treats synthetic data generation as mechanism design,...
Deep Dives
ReasoningBank: Giving Agents a Memory That Actually Learns from Mistakes
Google's ReasoningBank framework lets AI agents learn from both successes and failures after deployment, distilling...